game-ui-livestream
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructions require the agent to use browser, image, or video tools to visit and inspect content from
goodcase.aiandmedia.goodcase.ai. These are third-party domains not recognized as trusted vendors or well-known services. The agent is directed to read evidence and inspect media before producing artifacts. - [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external websites to guide its prompt generation logic, creating a vulnerability surface where instructions embedded in those pages or media files could influence the agent.
- Ingestion points: External URLs listed in
references/cases.mdand the output of tools used to inspect those links. - Boundary markers: Absent; the instructions do not provide specific delimiters or warnings to ignore potentially malicious instructions embedded in the external content.
- Capability inventory: The agent is authorized to use browser, image, and video tools to fetch and process external data.
- Sanitization: Absent; the skill does not specify any validation, filtering, or escaping of the content retrieved from the external sites before it is used to generate prompts or workflows.
Audit Metadata